[gcc]

2015-01-06  Bill Schmidt  <wschmidt@linux.vnet.ibm.com>

	* config/rs6000/vsx.md (*p9_vecload_<mode>): Replace VSX_M
	mode iterator with VSX_M2.
	(*p9_vecstore_<mode>): Likewise.
	(*vsx_le_permute_<mode>): Restrict to !TARGET_P9_VECTOR.
	(*vsx_le_perm_load_<mode> for VSX_LE_128): Likewise.
	(*vsx_le_perm_store_<mode> for VSX_LE_128): Likewise.
	(define_split for VSX_LE128 stores): Likewise.
	(define_peephole2 for TImode LE swaps): Likewise.
	(define_split for VSX_LE128 post-reload stores): Likewise.

[gcc/testsuite]

2015-01-06  Bill Schmidt  <wschmidt@linux.vnet.ibm.com>

	* gcc.target/powerpc/p9-lxvx-stxvx-3.c: New test.



git-svn-id: https://gcc.gnu.org/svn/gcc/trunk@232109 138bc75d-0d04-0410-961f-82ee72b054a4

4 files changed, 57 insertions, 11 deletions

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index e5b31af4893..eaae4ec69e1 100644
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,15 @@
+2015-01-06  Bill Schmidt  <wschmidt@linux.vnet.ibm.com>
+
+	* config/rs6000/vsx.md (*p9_vecload_<mode>): Replace VSX_M
+	mode iterator with VSX_M2.
+	(*p9_vecstore_<mode>): Likewise.
+	(*vsx_le_permute_<mode>): Restrict to !TARGET_P9_VECTOR.
+	(*vsx_le_perm_load_<mode> for VSX_LE_128): Likewise.
+	(*vsx_le_perm_store_<mode> for VSX_LE_128): Likewise.
+	(define_split for VSX_LE128 stores): Likewise.
+	(define_peephole2 for TImode LE swaps): Likewise.
+	(define_split for VSX_LE128 post-reload stores): Likewise.
+
 2016-01-06  Marek Polacek  <polacek@redhat.com>
 
 	PR sanitizer/69099
diff --git a/gcc/config/rs6000/vsx.md b/gcc/config/rs6000/vsx.md
index 4606bff208f..997ff31aef3 100644
--- a/gcc/config/rs6000/vsx.md
+++ b/gcc/config/rs6000/vsx.md
@@ -304,16 +304,16 @@
 ;; VSX (P9) moves
 
 (define_insn "*p9_vecload_<mode>"
-  [(set (match_operand:VSX_M 0 "vsx_register_operand" "=<VSa>")
-        (match_operand:VSX_M 1 "memory_operand" "Z"))]
+  [(set (match_operand:VSX_M2 0 "vsx_register_operand" "=<VSa>")
+        (match_operand:VSX_M2 1 "memory_operand" "Z"))]
   "TARGET_P9_VECTOR"
   "lxvx %x0,%y1"
   [(set_attr "type" "vecload")
    (set_attr "length" "4")])
 
 (define_insn "*p9_vecstore_<mode>"
-  [(set (match_operand:VSX_M 0 "memory_operand" "=Z")
-        (match_operand:VSX_M 1 "vsx_register_operand" "<VSa>"))]
+  [(set (match_operand:VSX_M2 0 "memory_operand" "=Z")
+        (match_operand:VSX_M2 1 "vsx_register_operand" "<VSa>"))]
   "TARGET_P9_VECTOR"
   "stxvx %x1,%y0"
   [(set_attr "type" "vecstore")
@@ -680,7 +680,7 @@
 	(rotate:VSX_LE_128
 	 (match_operand:VSX_LE_128 1 "input_operand" "<VSa>,Z,<VSa>")
 	 (const_int 64)))]
-  "!BYTES_BIG_ENDIAN && TARGET_VSX"
+  "!BYTES_BIG_ENDIAN && TARGET_VSX && !TARGET_P9_VECTOR"
   "@
    xxpermdi %x0,%x1,%x1,2
    lxvd2x %x0,%y1
@@ -714,9 +714,9 @@
 (define_insn_and_split "*vsx_le_perm_load_<mode>"
   [(set (match_operand:VSX_LE_128 0 "vsx_register_operand" "=<VSa>")
         (match_operand:VSX_LE_128 1 "memory_operand" "Z"))]
-  "!BYTES_BIG_ENDIAN && TARGET_VSX"
+  "!BYTES_BIG_ENDIAN && TARGET_VSX && !TARGET_P9_VECTOR"
   "#"
-  "!BYTES_BIG_ENDIAN && TARGET_VSX"
+  "!BYTES_BIG_ENDIAN && TARGET_VSX && !TARGET_P9_VECTOR"
   [(set (match_dup 2)
 	(rotate:VSX_LE_128 (match_dup 1)
 			   (const_int 64)))
@@ -735,7 +735,7 @@
 (define_insn "*vsx_le_perm_store_<mode>"
   [(set (match_operand:VSX_LE_128 0 "memory_operand" "=Z")
         (match_operand:VSX_LE_128 1 "vsx_register_operand" "+<VSa>"))]
-  "!BYTES_BIG_ENDIAN && TARGET_VSX"
+  "!BYTES_BIG_ENDIAN && TARGET_VSX && !TARGET_P9_VECTOR"
   "#"
   [(set_attr "type" "vecstore")
    (set_attr "length" "12")])
@@ -743,7 +743,7 @@
 (define_split
   [(set (match_operand:VSX_LE_128 0 "memory_operand" "")
         (match_operand:VSX_LE_128 1 "vsx_register_operand" ""))]
-  "!BYTES_BIG_ENDIAN && TARGET_VSX && !reload_completed"
+  "!BYTES_BIG_ENDIAN && TARGET_VSX && !reload_completed && !TARGET_P9_VECTOR"
   [(set (match_dup 2)
 	(rotate:VSX_LE_128 (match_dup 1)
 			   (const_int 64)))
@@ -765,7 +765,7 @@
    (set (match_operand:TI 2 "vsx_register_operand" "")
 	(rotate:TI (match_dup 0)
 		   (const_int 64)))]
-  "!BYTES_BIG_ENDIAN && TARGET_VSX && TARGET_VSX_TIMODE
+  "!BYTES_BIG_ENDIAN && TARGET_VSX && TARGET_VSX_TIMODE && !TARGET_P9_VECTOR
    && (rtx_equal_p (operands[0], operands[2])
        || peep2_reg_dead_p (2, operands[0]))"
    [(set (match_dup 2) (match_dup 1))])
@@ -775,7 +775,7 @@
 (define_split
   [(set (match_operand:VSX_LE_128 0 "memory_operand" "")
         (match_operand:VSX_LE_128 1 "vsx_register_operand" ""))]
-  "!BYTES_BIG_ENDIAN && TARGET_VSX && reload_completed"
+  "!BYTES_BIG_ENDIAN && TARGET_VSX && reload_completed && !TARGET_P9_VECTOR"
   [(set (match_dup 1)
 	(rotate:VSX_LE_128 (match_dup 1)
 			   (const_int 64)))
diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index 84581f7b524..ad953153de7 100644
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,3 +1,7 @@
+2015-01-06  Bill Schmidt  <wschmidt@linux.vnet.ibm.com>
+
+	* gcc.target/powerpc/p9-lxvx-stxvx-3.c: New test.
+
 2016-01-06  Andreas Schwab  <schwab@linux-m68k.org>
 
 	* gcc.dg/atomic-invalid-2.c: Remove duplicated content.
diff --git a/gcc/testsuite/gcc.target/powerpc/p9-lxvx-stxvx-3.c b/gcc/testsuite/gcc.target/powerpc/p9-lxvx-stxvx-3.c
new file mode 100644
index 00000000000..b298173b131
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/p9-lxvx-stxvx-3.c
@@ -0,0 +1,30 @@
+/* { dg-do compile { target { powerpc64le-*-* } } } */
+/* { dg-skip-if "do not override -mcpu" { powerpc*-*-* } { "-mcpu=*" } { "-mcpu=power9" } } */
+/* { dg-options "-mcpu=power9 -O3" } */
+/* { dg-final { scan-assembler "lxvx" } } */
+/* { dg-final { scan-assembler "stxvx" } } */
+/* { dg-final { scan-assembler-not "lxvd2x" } } */
+/* { dg-final { scan-assembler-not "stxvd2x" } } */
+/* { dg-final { scan-assembler-not "xxpermdi" } } */
+
+/* Verify P9 vector loads and stores are used rather than the
+   load-swap/swap-store workarounds for P8.  */
+#define SIZE (16384/sizeof(__float128))
+
+static __float128 x[SIZE] __attribute__ ((aligned (16)));
+static __float128 y[SIZE] __attribute__ ((aligned (16)));
+static __float128 a;
+
+void obfuscate(void *a, ...);
+
+void __attribute__((noinline)) do_one(void)
+{
+  unsigned long i;
+
+  obfuscate(x, y, &a);
+
+  for (i = 0; i < SIZE; i++)
+    y[i] = a * x[i];
+
+  obfuscate(x, y, &a);
+}